Molecular Chemistry-Controlled Hybrid Ink-Derived Efficient

Subscriber access provided by University of Otago Library

Letter

Molecular Chemistry Controlled Hybrid-ink Derived Efficient Cu2ZnSnS4 Photocathodes for Photoelectrochemical Water Splitting Wooseok Yang, Yunjung Oh, Jimin Kim, Myung Jin Jeong, Jong Hyeok Park, and Jooho Moon ACS Energy Lett., Just Accepted Manuscript • DOI: 10.1021/acsenergylett.6b00453 • Publication Date (Web): 07 Nov 2016 Downloaded from http://pubs.acs.org on November 9, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Energy Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

Molecular Chemistry Controlled Hybrid-Ink Derived Efficient Cu2ZnSnS4 Photocathodes for Photoelectrochemical Water Splitting

Wooseok Yang1, Yunjung Oh1, Jimin Kim1, Myung Jin Jeong2, Jong Hyeok Park2 and Jooho Moon1* 1

Department of Materials Science and Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, 120-749, Republic of Korea

2

Department of Chemical and Biomolecular Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, 120-749, Republic of Korea

*Corresponding author E-mail: [email protected], tel.: +82-2-2123-2855, fax: +82-2-312-5375.

1 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

ABSTRACT To realize economically competitive hydrogen production through photoelectrochemical (PEC) water splitting, it is essential to develop an efficient photoelectrode consisting of earthabundant constituents in conjunction with low-cost solution processing. Cu2ZnSnS4 (CZTS) has received significant attention as a promising photocathode owing to its abundance and good absorption properties. However, the efficiency of the solution-processed CZTS photocathode is not yet comparable to its counterparts. Here, a hybrid ink, obtained by careful control of precursor mixing order, was used to produce a highly efficient CZTS photocathode. The molecular chemistry controlled hybrid ink formulation, particularly the roles of thiourea–Sn2+ complexation, was elucidated by liquid Raman spectroscopy. The hybrid ink-derived CZTS thin films modified with conformal coating of an n-type TiO2/CdS double layer and a Pt electrocatalyst achieved an exceptionally high photocurrent of 13 mA cm-2 at −0.2 V vs. a reversible hydrogen electrode (RHE) under 1 Sun illumination. The modified photocathodes showed relatively stable H2 production with the faradaic efficiency close to unity.


Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

Table of Contents Graphic and Synopsis

A highly efficient Cu2ZnSnS4-based photocathode for water splitting is fabricated with wellcontrolled homogenous hybrid ink containing ionic Cu, Sn, thiourea, and metallic Zn nanoparticles. By sequential surface modifications with CdS, TiO2 and Pt, remarkable photocurrent of 13 mA cm-2 is achieved.


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

Realization of sustainable energy economy is contingent on developing technology that enables the conversion of renewable resources into an energy vector suitable for storage, transportation, and consumption. In the recently released Hydrogen and Fuel Cells

Technology Roadmap from the International Energy Agency, hydrogen is highlighted as an attractive energy carrier for this purpose, as it can utilize the current infrastructure for fossil fuels that is already in place.1 Moreover, hydrogen is advantageous for facilitating significant reductions in energy-related CO2 emissions with its low-carbon footprint. Semiconductorbased photoelectrochemical (PEC) water splitting is the one of the most promising technology for generating hydrogen by harnessing the enormous amount of renewable energy delivered from the Sun to the Earth, in conjunction with photovoltaic coupled electrolysis and photocatalytic water splitting. Since the first observation of PEC water splitting by Fujishima and Honda,2 many research groups have sought to boost the solar-to-hydrogen (STH) conversion efficiency while ensuring cost-competitiveness. STH efficiencies >20% may be necessary for commercialization. Based on a recent calculation by the Lewis group,3 the maximum STH efficiency of 29.7% can be achieved using a dual absorber (D4-type) PEC tandem cell composed of a series-connected n-type photoanode and a p-type photocathode, which is the most efficient configuration for PEC water splitting. Extensive investigations have been carried out to obtain promising photoanode materials to explore diverse n-type semiconductors with a high photocurrent and stability. In contrast, only a few studies have been reported for p-type photocathodes for PEC water splitting. While Huang et al. reported a survey of the recent efforts in developing photocathodes for hydrogen production,4 it is still a challenge to obtain p-type semiconductors using low-cost materials and processes with high performance. Among the various options available for p-type photocathode PEC water splitting, pCu2O is an attractive candidate because of its appreciable performance and abundancy.5-8 4 ACS Paragon Plus Environment

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

However, p-Cu2O has poor stability in aqueous solutions because the redox potentials for monovalent copper oxide lie between the valence and conduction bands of Cu2O. This instability makes the hermetic thin protective layer indispensable, thereby raising the device complexity and manufacturing cost. At the same time, p-type chalcogenide materials such as

p-CuInxGa(1-x)Se2 (CIGSe), p-CuInS2, p-CuGaSe2, and p-Cu2ZnSnS4 (CZTS) have also received increasing interest.9-11 In particular, non-toxicity and abundance of CZTS make it a promising material for low-cost and large-scale PEC devices. In this regard, a co-evaporated Pt/TiO2/CdS/CZTS photocathode was demonstrated, revealing a high photocurrent density near the 8 mA cm-2 level.12 More scalable routes like sol-gel spin-coating, electrodeposition, doctor blading, and the electrophoretic deposition of nanocrystal ink have also been reported as alternatives to expensive vacuum-based deposition processes.13-16 However, the efficiencies of solution-processed CZTS photocathodes are substantially low, presumably because of the phase impurity and low crystallinity of solution-derived CZTS thin films. Recently, relatively efficient photocathodes based on electrodeposited CZTS were reported by Jiang et al., exhibiting a photocurrent density of 9.3 mA cm-2 at 0 V vs. a reversible hydrogen electrode (RHE).17 However, they focused on a simple demonstration of the PEC performance without a systematic study of the relationship between processing conditions and phase formation, and lacked an in-depth analysis of CZTS precursor inks. The all-soluble precursor route is generally vulnerable to contamination from impurities such as carbon and oxygen because of incomplete decomposition of soluble precursors. This inevitably involves significant shrinkage and leads to the formation of porous structures with a small grain size.18 In contrast, the CZTS nanocrystal dispersion method offers stoichiometric advantages and purity controls at a molecular scale. However, it is difficult to fabricate dense organic residue-free thick films from the dispersion of nanocrystals capped with stabilizing molecules using complex nanoparticle synthesis.19 A 5 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

slurry approach, in which all insoluble precursor particles are suspended in alcohol, also suffers from the difficulty of achieving a homogeneous dispersion of the multicomponent precursor particles.20 The hydrazine-based hybrid-ink approach is the most promising route for fabricating high-quality CZTS thin films because hydrazine is a carbon- and oxygen-free solvent medium in which metal chalcogenides could be dissolved.21, 22 In hydrazine-based hybrid methodologies, S, Se, and metal chalcogenides such as Cu2S and SnSe are dissolved with no additives, whereas ZnS remains as a particulate species.23 Despite the unique chemistry of the hydrazine-based approach enabling the fabrication of highly pure and precisely stoichiometry-controlled CZTS films, hydrazine is highly toxic and explosive for practical use. In contrast, we used a nontoxic hybrid ink-based approach to produce highquality CZTS thin films.24 Our nontoxic hybrid ink consisted of dissolved ionic Cu and Sn species and thiourea in the presence of a homogeneously dispersed metallic Zn nanopowder in 2-methoxyethanol. However, our previous study simply demonstrated a comparison between Zn and ZnS particles as a Zn-source in terms of phase formation and solar cell efficiencies without consideration of the molecular chemistry in hybrid-ink formulation. Understanding the molecular chemistry in synthesizing the precursor solution is crucial because the compositional homogeneity in the as-deposited precursor film is dependent upon the chemical interaction among the constituent species. During the annealing process (generally under a H2S atmosphere at 550–580 °C), the detrimental secondary phases can be generated due to the nonuniformity of the as-deposited films and/or the differences in the diffusion rates and reactivities among the metallic species. For example, a high diffusion rate for Cu and the affinity of Cu to react with sulfur led to the formation of copper sulfides on the surface and voids left by the Cu atoms. A significant Sn-loss could occur by the evaporation of SnS2 because of its volatile nature; thus, the preferential formation of SnS2 should be avoided.25 It is also desirable for Zn to occur in the metallic phase because Zn ionic species 6 ACS Paragon Plus Environment

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

can more easily form a stable ZnS secondary phase that could slow the CZTS phase transformation.24 In order to prevent the elemental loss and detrimental secondary phase while achieving a nontoxic solvent-based homogeneous hybrid ink, a clear understanding of the complexation reactions among Cu, Sn, and sulfur precursors prior to the addition of welldispersed Zn nanoparticles is necessary. Thiourea, which is a versatile metal-ion complexing agent, has been extensively used as both a sulfur source and a stabilizer to ensure the dissolution of metal components for the nontoxic CZTS precursor solution.16, 18, 26 The synthesis procedure for the solution containing thiourea and the metal cation should be carefully controlled, otherwise the interaction between thiourea and the metal cation occasionally leads to precipitation, which causes inhomogeneity in the solution. However, the interaction between thiourea and the metal components has not yet been elucidated. Here, the role of thiourea in synthesizing the homogeneous solution was investigated by liquid Raman analysis. This is the first study to reveal the interaction between thiourea and a metal cation in solution for CZTS fabrication by liquid Raman spectroscopy. Based on a clear understanding of the molecular-level interaction between the precursors, we successfully fabricated compositionally homogenous and phasepure CZTS thin films from the synthesized hybrid ink. The subsequent deposition of CdS and TiO2 on CZTS as well as surface decoration with Pt enabled us to demonstrate an unprecedentedly enhanced performance, reaching a record-high performance for a CZTSbased photocathode in terms of a high photocurrent density and onset potential. A hybrid CZTS ink was synthesized by sequentially dissolving thiourea (TU), tin, and copper precursors in 2-methoxyethanol (2ME), followed by planetary milling with metallic Zn nanoparticles, as shown in Figure 1a. Firstly, TU was mixed with 2ME, resulting in a colorless clear solution. Upon the addition of SnCl2·2H2O to the solution containing TU, the colorless solution gradually changed to a light yellow solution. However, only the 7 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

SnCl2·2H2O-dissolved solution was colorless (Figure 1b), implying that the TU–Sn complex formation induced the coloration of the solution. Adding CuCl2·2H2O to the TU-Sn solution resulted in a further color change from light to dark yellow without any precipitation. In contrast, Figure 1b–e shows that different mixing orders for the precursors produced a precipitate. Without TU, the white precipitate was generated by mixing SnCl2·2H2O and CuCl2·2H2O, regardless of the mixing order (Figure 1b, c). Adding CuCl2·2H2O into the TUcontaining 2ME solution also resulted in an insoluble precipitate and vice versa (Figure 1d, e). This experimental observation implies that the mixing order of the precursors is an important factor, and the complexation reaction between TU and Sn especially plays a critical role in the preparation of the homogeneous Cu–Sn–TU precursor solution without any significant segregation. A hybrid CZTS ink preparation was completed by adding Zn nanoparticles to the clear solution containing TU, Sn, and Cu species, followed by planetary milling for 3 h. A small amount of dispersing additive (BYK 2163) was added to the TU–Sn– Cu solution prior to adding the Zn nanoparticles to ensure a homogeneous dispersion. The dispersed heavy Zn nanoparticles settled slowly under the influence of gravitational forces, but returned to an initial well-dispersed state by simple agitation, implying that the dispersing agent successfully prevented the agglomeration of Zn nanoparticles. Liquid Raman analysis was performed to elucidate the role of the TU–Sn complex. The symmetric stretching vibration of C=S in TU (at 738 cm-1) was slightly shifted to lower wavenumber (735 cm-1) upon the addition of the Sn precursor (Figure 2a), whereas the C–N stretching vibration (Figure 2b) shifted in the opposite direction (from 1086 to 1093 cm-1). The NH2 bending modes (at 3196 and 3327 cm-1) exhibited negligible changes upon the addition of the metal precursors (Figure 2c), indicating a negligible influence of the TU– metal complexation on the bonding character of the hydrogen atoms in TU. For thioureametal complexation, two possible coordinations, either through bonding to the nitrogen or 8 ACS Paragon Plus Environment

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

through bonding to the sulfur of TU, have been reported.27, 28 The molecular structure of the TU molecule is shown in Figure 2d. The interaction between the metal ions and sulfur in TU is expected to induce an enhanced delocalization of the lone pair electrons, resulting in a higher electron density developed on the sulfur atom. The enhanced delocalization reduces the double bonding character of the carbon–sulfur bond, eventually leading to a shift toward negative wavenumbers for the C=S stretching mode, with a positive shift of the C–N stretching mode.29 Thus, the negative shift after the addition of Sn into the TU-containing solution implies the formation of a TU–Sn complex through the interaction between the Sn2+ ion and S in TU. Adding the Cu precursor into the TU–Sn solution largely shifts the C=S vibration to lower wavenumbers (at 725 cm-1), implying a stronger interaction between TU and Cu than between TU and Sn (Figure 2a). To completely formulate the CZTS hybrid ink, metallic Zn nanoparticles were added to the homogeneous TU–Sn–Cu solution by a planetary milling process. The Raman spectrum of the complete CZTS ink (i.e., the TU–Sn–Cu–Zn ink) was almost identical to that of the TU–Sn–Cu ink, suggesting that the Zn metal particles had no effect on the interaction between TU and the metal cations because of the insoluble solid phase. To confirm the state of metallic Zn particles in the hybrid ink, thermogravimetric and differential scanning calorimetry (TG/DSC) analyses were performed under an inert atmosphere for the hybrid ink, which was dried at 250 °C under an inert atmosphere (Figure S1, Supporting Information). A strong endothermic peak at ~420 °C is indicative of Zn melting. This explains that the Zn nanoparticles existed in a dispersed metallic state without interacting with the TU–Sn–Cu complexes. In the case of the TU solution containing a Cu precursor without any Sn ions (Figure 1d), which led to a precipitate, the C=S stretching mode appeared at the same position as the TU–Sn–Cu solution, indicating that the interaction between TU and Cu is the same for both solutions (Figure S2, Supporting Information). However, there was an 9 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

additional peak in the TU solution containing the Cu precursor at 667 cm-1, which could be attributed to the polymerization of the TU–Cu complex. Several studies attributed the additional peaks near 640–670 cm-1 to polymerization, such as metal–oxygen–metal bridges in a prepared gel formed during the polycondensation process.30 This additional peak was used as an indicator for the degree of polymerization because the Raman signal was enhanced by the formation of long-range order or periodicity. The Raman results provide a glimpse as to the importance of the precursor mixing order to prepare the homogeneous CZTS solution. Because Cu2+ strongly reacts with S in TU, adding Cu2+ ions directly into the TU-containing 2ME solution produces a significant precipitation by the polymerization of Cu2+–TU. Sn2+ ions also react with S in TU, but the interaction between Sn2+ and TU is weaker than that between Cu2+ and TU, as expressed by the smaller shift of the C=S vibration. The TU–Sn2+ complex possibly plays a role in hindering the polymerization of the Cu–TU complex. The TU–Sn–Cu ink needs to be prepared according to the proper mixing order of the precursors to ensure no premature precipitation, after which Zn metallic nanoparticles are well dispersed in the CZTS hybrid ink. This is the first observation of the role of the TU–Sn2+ complex in the synthesis of a homogeneous hybrid precursor solution for CZTS thin films. We fabricated CZTS thin films using our synthesized hybrid inks with the sequential steps of spin coating, air drying, and sulfurization under a H2S atmosphere. In our previous study on the fabrication of a CZTS nanorod array, the drying temperature strongly affected the phase purity and PEC performance by ensuring both sufficient decomposition of the metal–TU complexes and removal of carbon impurities without too much oxidation and elemental losses.31 We also performed phase and compositional analyses on the hybrid inkderived CZTS thin films after sulfurization at 560 °C for 30 min, with different drying temperatures of 150 °C, 200 °C, 300 °C, and 400 °C in air. Note that the effect of the 10 ACS Paragon Plus Environment

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

sulfurization condition had been investigated in our previous study24 of CZTS thin film solar cells. Figure 3 shows the X-ray diffraction (XRD) peaks before/after sulfurization as a function of the drying temperatures. For the as-dried over 200 °C samples (Figure 3a), broad CZTS peaks were observed at 28.5°, 47.3°, and 56.1°, reflecting the (112), (220), and (312) planes of kesterite CZTS (JCPDS 26-0575), respectively. No clear peaks were observed in the sample dried at 150 °C, except for the Mo substrate because insufficient decomposition of the TU–metal complexes impeded the formation of metal oxide or metal sulfide compounds. With the increasing drying temperatures, the peaks for binary oxides such as Cu2O and SnO2 appeared with broad CZTS peaks. For the sulfurized sample dried at 150 °C, carbon-related impurity peaks were observed because the unremoved carbon during the low-temperature drying process likely reacted with other components during the high-temperature sulfurization process. When the samples were dried at 200 °C, only the CZTS peaks were visible after sulfurization, but the intensities of the CZTS peaks diminished on increasing the drying temperatures above 200 °C. The XRD data indicate that the formation of binary oxides during the air-drying of the precursor films hindered CZTS crystallization during the sulfurization process. A compositional analysis for the sulfurized CZTS thin films was performed by energy dispersive X-ray spectroscopy (EDS) analysis, as shown in Figure S3 (Supporting Information). On increasing the drying temperatures, the S/(Cu+Zn+Sn) ratio and the measured deviation increased. This result suggests that significant metal cation losses and nonuniformity were induced by raising the air-drying temperature. In contrast, the carbon concentration decreased as the drying temperature increased. The EDS analysis revealed that the optimum air drying temperature was 200 °C, at which the sample had a Cu-poor and Znrich composition (Cu/(Zn+Sn) = 0.91, Zn/Sn = 1.4, S/(Cu+Zn+Sn) = 1.0). As shown in Figure S4 (Supporting Information), our CZTS thin film contained densely packed and well-


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

grown crystallites with grain sizes above 500 nm. The film thickness was about 1 µm, which was close to that of the other efficient CZTS photocathodes. To clarify the presence of the secondary phases in our CZTS thin film dried at 200 °C, Raman analysis was performed with two different excitation wavelengths (532 and 325 nm). In both cases, the excitation power was kept below 50 µW to avoid thermal effects during the acquisition of the spectra. With the 532 nm wavelength (Figure S5a, Supporting Information), the CZTS film exhibited a strong peak at 337 cm-1 with weak shoulder peaks at 287 and 375 cm-1, all of which are attributed to CZTS. There were no noticeable peaks assigned to the secondary phases of the binary or ternary sulfide such as Cu2SnS3, Cu2S, and ZnS. It is noteworthy that the ZnS secondary phase is usually indiscernible from CZTS because the main peak locations for both are similar and the sensitivity of ZnS at 532 nm is low. A very small amount of ZnS (Eg ~ 3.7 eV) can be detected using UV Raman (325 nm wavelength, E ~ 3.8 eV), because the intensity of the Raman scattering is greatly enhanced when the incident laser frequency is close in energy to an electronic transition of the compound.32 In the Raman spectra recorded at 325 nm (Figure S5b, Supporting Information), a strong ZnS peak near 347 cm-1 was observed with a relatively broad full width at half maximum (FWHM), implying that the main peak can be deconvoluted into CZTS and ZnS peaks. The relative intensity difference for the CZTS and ZnS peaks did not explain the quantitative comparison between CZTS and ZnS because the resonance Raman signal for ZnS was much higher than that for the normal Raman signal for CZTS. These Raman results indicate the minimal presence of the ZnS secondary phase, but its quantitative amount is still in question. The ZnS secondary phase is considered as an inevitable consequence in the Cupoor and Zn-rich composition, which is a common feature of even highly efficient CZTS solar cells, to develop shallow acceptors (VCu) and suppress deep acceptors (CuZn).33 As Berg

et al. demonstrated, the detection limits for the secondary phase in CZTS using XRD and 12 ACS Paragon Plus Environment

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

Raman spectroscopy were too low, further analysis tool is necessary to quantitatively discriminate the ZnS impurity from CZTS.34 Figure S6 (Supporting Information) shows the photocurrent density-potential curves of the bare CZTS thin films dried at different temperatures after sulfurization at 560 °C for 30 min. Samples were submerged in 0.5 M Na2SO4 aqueous solution (the pH was adjusted to 10 by the addition of NaOH) under simulated light illumination. All samples provided a cathodic photocurrent, indicative of the p-type semiconducting properties of our CZTS thin films. The CZTS film dried at 200 °C showed the largest photocurrent: 166 µA cm-2 at −0.2 V vs. RHE, which matches well to the phase and composition analysis results. The surface of the p-type photocathodes is commonly coated with thin layers of the n-type buffer layer to form a p–n junction. The resulting built-in potential, created across the p–n junction by equilibration of the Fermi level, induces a steeper band bending and widens the depletion width, facilitating the charge separation and shifting the onset potential.35 We deposited an n-type CdS layer using a chemical bath deposition (CBD) process followed by spin-coating a solutionprocessed TiO2 layer (s-TiO2). Pt has a very low overpotential for hydrogen evolution and was also deposited as an electrocatalyst on the n-type layer. Figure 4 shows the PEC properties of the CZTS photocathode with the deposition of

n-type layers, and the Pt catalyst under the same measurement conditions used for Figure S6 (Supporting Information). A significant enhancement of the photocurrent was achieved by the deposition of a CdS layer, as expected from the results for similar systems.12, 17 The thickness of CdS was optimized by controlling the CBD times, as shown in Figure S7 (Supporting Information). The influence of the CdS thickness on the p-type chalcopyrite thin films has been discussed elsewhere.11 The photocurrent density at −0.2 V vs. RHE of the Pt/CdS/CZTS electrode reached 2.2 mA cm-2, which is over 13 times higher than that of the CZTS electrode. It is noteworthy that no distinct enhancement in the photocurrent was observed 13 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

despite of high catalytic property of Pt catalyst (Figure S8). This observation suggests that the main limitation of CZTS photocathode relates to slow charge separation rather than low catalytic activity as discussed in literature.36 Using a s-TiO2/CdS double n-type layer, further enhancement of the photocurrent density was observed: the photocurrent density above 5.5 mA cm-2 at −0.2 V vs. RHE. In addition to improved photocurrent, a shift of the onset potential was also observed upon deposition of n-TiO2 layer. It is noteworthy that flat band potential can be estimated from the onset potential where a photocurrent arises.13 We define the onset potentials as the potential showing 50 µA cm-2 of cathodic photocurrent in the I–E curves according to literature.37 In a recent study of the efficient CZTS-based photocathode in the optimum Pt/In2S3/CdS/CZTS structure, the onset potentials of both Pt/In2S3/CdS/CZTS (0.63 V vs. RHE) electrode and Pt/CdS/CZTS (0.64 V vs. RHE) were nearly identical (i.e., the n-type In2S3 layer did not affect the onset potential despite the photocurrent improvement).17 In our case, the insertion of the n-TiO2 layer resulted in a significant shift of the onset potential to 0.86 V vs. RHE compared to the Pt/CdS/CZTS electrode (0.53 V vs. RHE). This result suggests that TiO2 was more efficient than In2S3, probably due to the larger band gap of TiO2 (~3.2 eV) than that of In2S3 (~2.24 eV). The deeper valence band of TiO2 is believed to induce a larger electrostatic repulsion of holes from the surface, thereby reducing surface recombinations.35 It should be noted that the determination of onset potential for bare CZTS sample would yield arbitrary value because of high off current level, indicating that the onset simply represents the decomposition potential. The large onset potential, implying large flat band potential, for our Pt/s-TiO2/CdS/CZTS photocathode is a record value for a CZTSbased photoelectrode, which is beneficial to develop zero-bias tandem cells. The schematic of band diagram depicting the energetics of the Pt/TiO2/CdS/CZTS photocathode before/after equilibrium in the electrolyte is presented in Figure S9. The flat band potentials of CZTS, CdS and TiO2 are assumed to be 0.65 V, −0.1 V and −0.2 V vs RHE, respectively.35,38 The 14 ACS Paragon Plus Environment

Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

fermi level and band alignment among CZTS, CdS, and TiO2 were reported in literature.35.39 In a schematic diagram, the equilibration would lead to form a type-II heterojunction where the electrons can selectively transfer from CZTS to the electrolyte through the multi-junction of CdS and TiO2. In contrast, the deposition of TiO2 directly on the CZTS film would result in deactivation of photoelectrochemical properties of CZTS because the interface between CZTS and TiO2 would act as a recombination center possibly due to large band gap difference, as discussed in the literature.38 To further boost the photocurrent of our hybrid ink-derived CZTS-based electrode, the PEC performance was evaluated in an aqueous mixed solution of 0.5 M Na2SO4, 0.25 M Na2HPO4, and 0.25 M NaH2PO4 (the pH was adjusted to 6.85 by NaOH addition) under simulated AM 1.5 G irradiation (Figure 5a). The photocurrent of the Pt/s-TiO2/CdS/CZTS electrode reached a remarkably high value of 21.5 mA cm-2 at −0.2 V vs. RHE. It is worth to note the presence of higher transient current spike at larger cathodic bias. The transient current spike results from the recombination of photo-generated carrier due to two possible origins. Firstly, interfacial trap states and/or sluggish charge transfer kinetics can cause recombination, so that the photocurrent spike is likely to disappear at larger cathodic bias. Secondly, the recombination due to the reduced species of the redox couple at the electrode/electrolyte interface can result in the transient current spike preferred at higher photocurrent region. In our case, the presence of photocurrent spike can be attributed to drastic consumption of redox couple at higher cathodic bias.13 Kumagai et al. elucidated the difference between the Na2SO4 and phosphate buffer solutions by obtaining current-potential curves of a rotating Pt disk electrode in both solutions. Through the Pt electrochemistry, they ascribed the origin of the difference to the variations in the hydrogen evolution reaction (HER) at the surface of the Pt particles and/or the role of the buffer species that stabilize the pH in the vicinity of the electrode surface, which reduces the chemical bias, thus promoting 15 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

the HER.9 Our results that the comparison e between two electrolytes are in good agreement with the previous study on the Pt/CdS/CIGSe photocathodes and the theory of near-surface ion distribution and buffer effect.9,

40

The amount of evolved H2 from the Pt/s-

TiO2/CdS/CZTS photocathode was evaluated in the three-electrode configuration at 0 V vs. RHE under AM 1.5G radiation by gas chromatography (GC). Figure S10 shows the typical time course curves for both H2 evolution from the photocathode with an active area of 0.16 cm2 and one-half of the electrons passing through the outer circuit (e-/2) as expressed by a solid line. Comparing the determined hydrogen generation with the expected amount calculated by the photocurrent, the faradaic efficiency (i.e., the ratio of the rate of H2 evolution to that of e-/2) can be calculated. Despite the exceptionally high photocurrent, the solution processed TiO2 based photocathode reveals 45-55 % of faradaic efficiency, indicating the presence of side reactions rather than H2 evolution, possibly due to an incomplete coverage of thin TiO2. Device stability is one of the most crucial features to realize commercially viable water splitting. Because CZTS and CdS are relatively unstable during the PEC measurement compared to TiO2, a conformally coated TiO2 layer is necessary to achieve stable PEC performance. The spin-coating process for the TiO2 layer was optimized based on linear sweep voltammograms and chronoamperometric measurements at 0 V vs. RHE (Figure S11). Despite the high photocurrent with a relatively thin TiO2 layer (single or double coating of 0.05 M TiO2 precursor solution), the photocurrent remained at only 15% of its initial value after 1 h. This instability presumably resulted from the incomplete coverage of the thin TiO2 layer on the CdS/CZTS surfaces. Although the relatively thick spin-coated TiO2 layer (double coating of a 0.1 M TiO2 solution) on the CZTS-based photocathode yielded a better stability (retained 55% of its initial value after 1 h), the photocurrent level was substantially


Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

diminished compared to the thin-TiO2 layer, likely due to the charge recombination and optical loss associated with a thick TiO2 layer. To achieve a better stable photocathode with a high photocurrent, we used atomic layer deposition (ALD) for the TiO2 layer, which is a desirable technique for the conformal coating of a thin layer. The thickness-dependent transmittance of ALD-TiO2 layer is presented in Figure S12. As expected, the transmittance of TiO2 layer decreases with the increasing thickness. We adopted 600 cycles ALD-TiO2 in order to avoid optical loss within visible light range (400 – 800 nm), which is close to absorption range of CZTS. Figure 6a shows the cross-sectional SEM image for the Pt/ALD-TiO2/CdS/CZTS electrode. Thin TiO2 and CdS were observed as conformal layers covering the underneath CZTS films. To further confirm the vertical distribution of the different layers in the photoelectrode, contour plots were constructed using X-ray photoelectron spectroscopy (XPS) depth profile analysis (Figure 6b–e). XPS contour plots clearly show the sequential stacks of Pt/TiO2/CdS/CZTS with distinct boundaries, implying conformal coating of the TiO2 and CdS layers and no exposure of CdS and CZTS to the electrolyte. The thickness of TiO2 and CdS were estimated to 25 and 35 nm, respectively. The current density-potential and current density-time curves are presented in Figure 7. The ALD-TiO2-deposited CZTS photocathode (Pt/ALDTiO2/CdS/CZTS) exhibited improved stability during the PEC measurement (75% retained photocurrent after 1 h) with a minor variation due to the accumulation/detachment of bubbles on the surface of the electrode adjacent to the insulating epoxy, as shown in Figure S13 (Supporting Information). Although the ALD-TiO2-deposited photocathode revealed better stability than its solution-deposited counterpart, the photocurrent of the Pt/ALDTiO2/CdS/CZTS photocathode decreased continuously during the long-term stability test: only 30% of the photocurrent was maintained after 5 h (Figure S14, Supporting Information). Paracchino et al. reported an enhanced stability of the Cu2O photocathode with 17 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

the protection of a double layer of ALD-TiO2 and ALD-TiO2-aluminum-doped ZnO, but the photocurrent of ALD-protected Cu2O decreased by 50% after only 20 min.5 They ascribed the instability to a possible reduction of TiO2 to Ti(OH)3. Rovelli et al. also observed 55% of the retaining photocurrent after 15 min for a similarly ALD-protected CZTS photocathode in the structure of ALD-TiO2/AZO/CdS/Mo. They claimed that the photocurrent drop was due to the mass-transfer limitation and bubble formation in the electrolyte side, rather than the instability of the protection layer. This explanation was supported by the fact that the photocurrent be recovered to some extent by several means such as by depositing an additional layer of Pt catalyst, by replacing the used electrolyte with a fresh solution, and even by simple agitation of the electrolyte.38 Similarly, we observed that the photocurrent drop of our CZTS-based photocathode was completely recovered after 12 h, indicating that the photocurrent drop could originate from the mass transfer-related issue and bubble formation. The amount of evolved H2 from the Pt/ALD-TiO2/CdS/CZTS photocathode was also evaluated. As shown in Figure 8, Pt/ALD-TiO2/CdS/CZTS photocathode revealed 90-95% of faradaic efficiency, indicating the absence of side reactions. Incident photon to current efficiency (IPCE) of the Pt/ALD-TIO2/CdS/CZTS photocathode was investigated under irradiation of monochromatic light with various wavelength at 0 V vs RHE. The photon flux was determined using a calibrated Si photodiode. The obtained IPCE spectra is shown Figure S15. By converting the spectra into [hν ln(1-IPCE)]2 versus photon energy (hν/eV) plot, the band gap energy (Eg) of the CZTS was estimated to be 1.53 eV, which is close to the literature.24 In summary, we demonstrated highly efficient photocathodes based on low-cost solution-processed CZTS thin films. To accomplish this, we developed a desirable synthetic route for homogeneous hybrid inks containing ionic Cu, Sn, and thiourea together with well18 ACS Paragon Plus Environment

Page 19 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

dispersed Zn nanoparticles, enabling a high-quality CZTS photocathode. Liquid Raman spectroscopy allowed us to understand the role of Sn-thiourea complexation in the preparation of the hybrid ink, which is capable of hindering the polymerization of the Cu ion– thiourea complex and in turn, the undesirable segregation in the precursor ink state. The airdrying temperature of the as-deposited hybrid ink-derived CZTS films played an important role in determining the phase evolution and crystallization. High drying temperatures above 300 °C resulted in nonuniformity and elemental loss, whereas low drying temperatures near 150 °C were unable to remove carbon impurities from the precursor films. The PEC performance of our hybrid ink-derived CZTS photocathode was greatly enhanced by the adoption of thin n-type overlayers and Pt co-catalyst, particularly in the phosphate buffer solution. Despite the exceptionally high photocurrent of solution-processed TiO2 deposited photocathode reaching 21.5 mAcm-2, the faradaic efficiency is within a range of 45-55 %, indicating the presence of side reaction possibly due to an incomplete coverage. For better stability, an ALD-TiO2 layer was used, and the ALD-TiO2-deposited photocathode retained 75% of its initial photocurrent after 1 h of measurement time at 0 V vs. RHE. GC was used to detect the hydrogen evolved from the CZTS-based photocathode, and the faradaic efficiency reached nearly 100%. Our hybrid ink-derived CZTS photocathodes clearly demonstrated a great

potential

in

solar

water

splitting

for

viable

hydrogen

production

via

photoelectrochemical tandem cells.


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 32

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/xx. Experimental section and Figures S1−S15 are provided: TG and DSC of the dried CZTS hybrid ink; Raman spectra for thiourea and the Cu–thiourea solution; Relative composition ratio of each element and carbon concentration (%); SEM images of a hybrid ink-derived CZTS thin film. Raman scattering spectra with 532 nm and 325 nm excitation wavelengths; Current density-potential curves of CZTS, Pt/CZTS, Pt/CdS/CZTS thin films; Schematics of band diagram; H2 evolution over the Pt/solution-TiO2/CdS/CZTS; Current density-potential curves of a Pt/s-TiO2/CdS/CZTS; Thickness dependent transmittance of ALD-TiO2 films; Photographs of PEC device; Stability test (5 h) by the current density-time curve; IPCE spectrum of Pt/ALT-TiO2/CdS/CZTS photocathode.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This work was supported by a National Research Foundation (NRF) of Korea grant funded by the Korean government (MSIP) (No. 2012R1A3A2026417). 20 ACS Paragon Plus Environment

Page 21 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

REFERENCES (1) Miller, E. L. Photoelectrochemical Water Splitting. Energy Environ. Sci. 2015, 8, 28092810. (2) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37-38. (3) Hu, S.; Xiang, C.; Haussener, S.; Berger, A. D.; Lewis, N. S. An Analysis of the Optimal Band Gaps of Light Absorbers in Integrated Tandem Photoelectrochemical Water-splitting Systems. Energy Environ. Sci. 2013, 6, 2984-2993. (4) Huang, Q.; Ye, Z.; Xiao, X. Recent Progress in Photocathodes for Hydrogen Evolution. J. Mater. Chem. A 2015, 3, 15824-15837. (5) Paracchino, A.; Laporte, V.; Sivula, K.; Grätzel, M.; Thimsen, E. Highly Active Oxide Photocathode for Photoelectrochemical Water Reduction. Nat. Mater. 2011, 10, 456-461. (6) Morales-Guio, C. G.; Tilley, S. D.; Vrubel, H.; Grätzel, M.; Hu, X. Hydrogen Evolution from a Copper(I) Oxide Photocathode Coated with an Amorphous Molybdenum Sulphide Catalyst. Nat. Commun. 2014, 5, 3059. (7) Dias, P.; Schreier, M.; Tilley, S. D.; Luo, J.; Azevedo, J.; Andrade, L.; Bi, D.; Hagfeldt, A.; Mendes, A.; Grätzel, M. Transparent Cuprous Oxide Photocathode Enabling a Stacked Tandem Cell for Unbiased Water Splitting. Adv. Energy Mater. 2015, 5, 1501537. (8) Li, C.; Hisatomi, T.; Watanabe, O.; Nakabayashi, M.; Shibata, N.; Domen, K.; Delaunay, J.-J. Positive Onset Potential and Stability of Cu2O-based Photocathodes in Water Splitting by Atomic Layer Deposition of a Ga2O3 Buffer Layer. Energy Environ. Sci. 2015, 8, 14931500. (9) Kumagai, H.; Minegishi, T.; Sato, N.; Yamada, T.; Kubota, J.; Domen, K. Efficient Solar Hydrogen Production from Neutral Electrolytes using Surface-modified Cu(In,Ga)Se2 Photocathodes. J. Mater. Chem. A 2015, 3, 8300-8307. (10) Septina, W.; Ikeda, S.; Harada, T.; Minegishi, T.; Domen, K.; Matsumura, M. Platinum and Indium Sulfide-modified CuInS2 as Efficient Photocathodes for Photoelectrochemical Water Splitting. Chem. Commun. 2014, 50, 8941-8943. (11) Moriya, M.; Minegishi, T.; Kumagai, H.; Katayama, M.; Kubota, J.; Domen, K. Stable Hydrogen Evolution from CdS-Modified CuGaSe2 Photoelectrode under Visible-Light Irradiation. J. Am. Chem. Soc. 2013, 135, 3733-3735.


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 32

(12) Yokoyama, D.; Minegishi, T.; Jimbo, K.; Hisatomi, T.; Ma, G.; Katayama, M.; Kubota, J.; Katagiri, H.; Domen, K. H2 Evolution from Water on Modified Cu2ZnSnS4 Photoelectrode under Solar Light. Appl. Phys. Express 2010, 3, 101202. (13) Guijarro, N. s.; Prévot, M. S.; Sivula, K. Enhancing the Charge Separation in Nanocrystalline Cu2ZnSnS4 Photocathodes for Photoelectrochemical Application: the Role of Surface Modifications. J. Phys. Chem. Lett. 2014, 5, 3902-3908. (14) Guan, Z.; Luo, W.; Zou, Z. Formation Mechanism of ZnS Impurities and Their Effect on Photoelectrochemical Properties on a Cu2ZnSnS4 Photocathode. CrystEngComm 2014, 16, 2929-2936. (15) Zhang, Y.; Ouyang, S.; Yu, Q.; Li, P.; Ye, J. Modulation of Sulfur Partial Pressure in Sulfurization to Significantly Improve the Photoelectrochemical Performance over the Cu2ZnSnS4 Photocathode. Chem. Commun. 2015, 51, 14057-14059. (16) Guan, Z.; Luo, W.; Xu, Y.; Tao, Q.; Wen, X.; Zou, Z. Aging Precursor Solution in High Humidity Remarkably Promoted Grain Growth in Cu2ZnSnS4 Films. ACS Appl. Mater. Interfaces 2016, 8, 5432-5438. (17) Jiang, F.; Harada, T.; Kuang, Y.; Minegishi, T.; Domen, K.; Ikeda, S. Pt/In2S3/CdS/Cu2ZnSnS4 Thin Film as an Efficient and Stable Photocathode for Water Reduction under Sunlight Radiation. J. Am. Chem. Soc. 2015, 137, 13691-13697. (18) Chaudhuri, T. K.; Tiwari, D. Earth-abundant Non-toxic Cu2ZnSnS4 Thin Films by Direct Liquid Coating from Metal–thiourea Precursor Solution. Sol. Energy Mat. Sol. Cells 2012, 101, 46-50. (19) Kim, Y.; Woo, K.; Kim, I.; Cho, Y. S.; Jeong, S.; Moon, J. Highly Concentrated Synthesis of Copper-zinc-tin-sulfide Nanocrystals with Easily Decomposable Capping Molecules for Printed Photovoltaic Applications. Nanoscale 2013, 5, 10183-10188. (20) Woo, K.; Kim, Y.; Moon, J. A Non-toxic, Solution-processed, Earth Abundant Absorbing Layer for Thin-film Solar Cells. Energy Environ. Sci. 2012, 5, 5340-5345. (21) Todorov, T. K.; Reuter, K. B.; Mitzi, D. B. High-Efficiency Solar Cell with EarthAbundant Liquid-Processed Absorber. Adv. Energy Mater. 2010, 22, E156-E159 (22) Todorov, T. K.; Tang, J.; Bag, S.; Gunawan, O.; Gokmen, T.; Zhu, Y.; Mitzi, D. B. Beyond 11% Efficiency: Characteristics of State-of-the-Art Cu2ZnSn(S,Se)4 Solar Cells. Adv. Energy Mater. 2013, 3, 34-38. (23) Bag, S.; Gunawan, O.; Gokmen, T.; Zhu, Y.; Todorov, T. K.; Mitzi, D. B. Low Band Gap Liquid-Processed CZTSe Solar Cell with 10.1% Efficiency. Energy Environ. Sci. 2012, 5, 7060-7065.


Page 23 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

(24) Kim, K.; Kim, I.; Oh, Y.; Lee, D.; Woo, K.; Jeong, S.; Moon, J. Influence of Precursor Type on Non-toxic Hybrid Inks for High-Efficiency Cu2ZnSnS4 Thin-film Solar Cells. Green Chem. 2014, 16, 4323-4332. (25) Kumar, M.; Dubey, A.; Adhikari, N.; Venkatesan, S.; Qiao, Q. Strategic Review of Secondary Phases, Defects and Defect-Complexes in Kesterite CZTS–Se Solar Cells. Energy Environ. Sci. 2015, 8, 3134-3159. (26) Xin, H.; Katahara, J. K.; Braly, I. L.; Hillhouse, H. W. 8% Effi cient Cu2ZnSn(S,Se)4 Solar Cells from Redox Equilibrated Simple Precursors in DMSO. Adv. Energy Mater. 2014, 4, 1301823 (27) Kannan, V.; Rajesh, N.; Ganesh, R. B.; Ramasamy, P. Growth and Characterization of Bisthiourea-Zinc Acetate, a New Nonlinear Optical Material. J. Cryst. Growth 2004, 269, 565-569. (28) Selvakumar, S.; Kumar, S. R.; Joseph, G. P.; Rajarajan, K.; Madhavan, J.; Rajasekar, S.; Sagayaraj, P. Growth and Characterization of Pure and Doped Bis(thiourea) Cadmium Acetate Single Crystals. Mater. Chem. Phys. 2007, 103, 153-157. (29) Kumari, R. G.; Ramakrishnan, V.; Carolin, M. L.; Kumar, J.; Sarua, A.; Kuball, M. Raman Spectral Investigation of Thiourea Complexes. Spectrochim. Acta Part A: Molecular and Biomolecular Spectroscopy 2009, 73, 263-267. (30) Granqvist, C. G., Handbook of Inorganic Electrochromic Materials, Elsevier Science, Amsterdam, Netherlands; 1995. (31) Kim, J.; Yang, W.; Oh, Y.; Kim, J.; Moon, J.; Template-directed Fabrication of Vertically Aligned Cu2ZnSnS4 Nanorod Arrays for Photoelectrochemical Applications via a Non-toxic Solution Process. J. Alloy. Compd. 2017, 691, 457-465. (32) Fontané, X.; Calvo-Barrio, L.; Izquierdo-Roca, V.; Saucedo, E.; Pérez-Rodriguez, A.; Morante, J.; Berg, D.; Dale, P.; Siebentritt, S. In-depth Resolved Raman Scattering Analysis for the Identification of Secondary Phases: Characterization of Cu2ZnSnS4 Layers for Solar Cell Applications. Appl. Phys. Lett. 2011, 98, 181905. (33) Chen, S.; Walsh, A.; Gong, X. G.; Wei, S. H. Classification of Lattice Defects in the Kesterite Cu2ZnSnS4 and Cu2ZnSnSe4 Earth-Abundant Solar Cell Absorbers. Adv. Mater. 2013, 25, 1522-1539. (34) Berg, D. M.; Arasimowicz, M.; Djemour, R.; Gütay, L.; Siebentritt, S.; Schorr, S.; Fontané, X.; Izquierdo-Roca, V.; Pérez-Rodriguez, A.; Dale, P. J. Discrimination and Detection Limits of Secondary Phases in Cu2ZnSnS4 Using X-ray Diffraction and Raman Spectroscopy. Thin Solid Films 2014, 569, 113-123. (35) Guijarro, N.; Prévot, M. S.; Yu, X.; Jeanbourquin, X. A.; Bornoz, P.; Bourée, W.; Johnson, M.; Le Formal, F.; Sivula, K. A Bottom-Up Approach toward All-Solution23 ACS Paragon Plus Environment

ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 32

Processed High-Effi ciency Cu(In,Ga)S2 Photocathodes for Solar Water Splitting. Adv. Energy Mater. 2016, 6, 1501949. (36) Sivula, K.; van de Krol, R.; Semiconducting Materials for Photoelectrochemical Energy Conversion. Nat. Rev. Mater., 2016, 1, 15010. (37) Zhang, L.; Minegishi, T.; Kubota, J.; Domen, K. Hydrogen Evolution from Water using AgxCu1-xGaSe2 Photocathodes under Visible Light. Phys. Chem. Chem. Phys. 2014, 16, 6167-6174. (38) Rovelli, L.; Tilley, S. D.; Sivula, K. Optimization and Stabilization of Electrodeposited Cu2ZnSnS4 Photocathodes for Solar Water Reduction. ACS Appl. Mater. interfaces 2013, 5, 8018-8024. (39) Huang, S.; Luo, W.; Zou, Z. Band Positions and Photoelectrochemical Properties of Cu2ZnSnS4 Thin Films by the Ultrasonic Spray Pyrolysis Method. J. Phys. D: Appl. Phys. 2013, 46, 235108. (40) Auinger, M.; Katsounaros, I.; Meier, J. C.; Klemm, S. O.; Biedermann, P., U.; Topalov, A. A.; Rohwerder, M.; Mayrhofer, K. J. J. Near-Surface Ion Distribution and Buffer Effects During Electrochemical Reactions. Phys. Chem. Chem. Phys., 2011, 13, 16384-16394.


Page 25 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

Figure 1. Photographs of (a) the desirable route for synthesizing the homogeneous CZTS hybrid ink containing ionic TU–Sn–Cu complexes with well-dispersed metallic Zn nanoparticles, (b)/(c) the absence of the TU complexing agent results in precipitation, and (d)/(e) undesirable mixing orders of the precursors also produces the precipitate.


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 32

Figure 2. Raman spectra corresponding to (a) C=S stretching, (b) C–N stretching, and (c) NH2 bending for CZTS hybrid inks with sequential addition of each precursor. (d) Schematics of the molecular structure of thiourea before/after bonding with the metal cations.


Page 27 of 32

(a)

(b)

150 C dried o 200 C dried o 300 C dried o 400 C dried

Intensity (a.u.)

150 C dried o 200 C dried o 300 C dried o 400 C dried

Mo

20

30

40

o

o

Mo

CZTS Cu2O SnO2 Intensity (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

50

60

2Theta (degree)

70

80

CCu2O5

20

30

CCu2O5

CCu2O5

40

50

60

70

80

2Theta (degree)

Figure 3. XRD patterns of (a) the as-dried and (b) after sulfurized hybrid ink-derived CZTS films as a function of drying temperatures ranging from 150 °C to 400 °C. The sulfurization condition was fixed at 560 °C under a H2S atmosphere for 30 min.


ACS Energy Letters

CZTS x10 Off Pt/CdS/CZTS

2

Current density (mA/cm )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 32

On Off Pt/s-TiO2/CdS/CZTS

On

3 mA/cm2

-0.2

0.0

0.2

0.4

0.6

0.8

Potential / V vs. RHE

Figure 4. Current density-potential curves of the CZTS, Pt/CdS/CZTS, and Pt/sTiO2/CdS/CZTS photocathodes in a 0.5 M Na2SO4 aqueous solution (pH adjusted to 10 by the addition of NaOH) under chopped solar-simulated light illumination (AM 1.5G). The sTiO2 layer was deposited by an ethanol-based solution process. The scan rate was 5 mV/s with a negative scan direction.


Page 29 of 32

CZTS x10 Off Pt/CdS/CZTS 2

Current density (mA/cm )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

On Off Pt/s-TiO2/CdS/CZTS

10 mA/cm2 On

-0.2

0.0

0.2

0.4

0.6

0.8


Figure 5. Current density-potential curves of the CZTS, Pt/CdS/CZTS, and Pt/sTiO2/CdS/CZTS photocathodes in a phosphate-buffered aqueous solution with pH 6.85 under chopped solar-simulated light illumination (AM 1.5G). The scan rate was 5 mV/s with a negative scan direction.


ACS Energy Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 32

Figure 6. (a) Cross-sectional SEM image of a Pt/ALD-TiO2/CdS/CZTS photocathode. (b–e) The corresponding elemental contour plots for the Pt, Ti, Cd, and Cu XPS signals.


Page 31 of 32

(b)

0

Current density (mA/cm2)

(a) Current density (mA/cm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

-5

-10

-15 -0.2

Pt/ALD-TiO2/CdS/CZTS 0.0

0.2

0.4

0.6

0.8

0 -2 -4

Bubble detachment

-6 -8 -10 -12

0

600

1200


1800

2400

3000

3600

Time (s)

Figure 7. (a) Current density-potential curves of a Pt/ALD-TiO2/CdS/CZTS photocathode in a phosphate-buffered aqueous solution at pH 6.85 under chopped solar-simulated light illumination (AM 1.5G) and (b) the corresponding current density-time curve with an applied bias of 0 V vs. RHE.


ACS Energy Letters

25

Light off H2 production (µmol)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 32

20

e-/2

15

H2

10

5 0

0

20

40

60

80

100

Time (min) Figure 8. Time course curve of H2 evolution over the Pt/ALD-TiO2/CdS/CZTS photocathode under simulated sunlight (AM 1.5G) at 0 V vs. RHE. The solid line denotes the time course curve for one-half of the electrons passing through the outer circuit (e-/2).


Molecular Chemistry-Controlled Hybrid Ink-Derived Efficient

Recommend Documents